Patterns of Peripheral Formation of Ankara and Types of Lost Space

40 these housing developments ILKO and Park Street and environs were integrated as partial plans (Kızıltaş, 2010). The Park Street and environs also were aimed to make a buffer zone between Çayyolu and Alacaatlı district.

The third district is made of Yaşamkent and Alacaatlı subdistricts. The first developments date back after 1985 where expansions in this area have been based on partial plans. Yaşamkent is an example of such approach. The preparations and implementations of the plan have been prepared and implemented according to different stages, therefore modifications have occured through time (Kızıltaş, 2010). Alacaatlı villages which is as well a subdistrict is based on partial plan development and its boundaries extend further until ‘Çayyolu Mass Housing’ area and aims a homogeneous density around the area (Kızıltaş, 2010).

41 The reason behind the lack of variety in urban patterns of periphery is the design method itself, where morphological configurations are determined by the border definition of roads as the main factor. Çalışkan (2012) analyzed the morphology of residential units in the peripheries of Ankara in five different levels to find differences and similarities: plot layout, street pattern, pedestrian network, figure-ground relationship and massing. According to Çalışkan (2012), coherence is a condition of morphological unity, where all essential elements interact with each other and are integrated to form modules of a higher level. If such a morphological condition is met, then spatial configuration of urban and architectural elements would be well achieved.

In the analysis of the pattern formation, Çalışkan found out that the main problem with coherence in all layers is that they do not provide the sense of an interior space that can be seen as the interface of building unit integration. In addition, the massing layer has a weak coherence which is the lack of diversity in building units and sloppy connections between large and smaller units (Çalışkan, 2012). A huge number of sampling sites are composed of same reparative unit type which results in a lack of complexity for systematic entirety. Since development plans in Turkey and Ankara specifically, have always been associated with partial plans or modifications of smaller scales, this has resulted in designs that at times have not taken into consideration outer reference to the site morphologically within the context of urban fabric. Alexander (1987), prefers to call this fragmented spatial structure as ‘piecemeal growth’, which for him is incoherent and scattered, and can produce assemblies rather than coherent wholes if powerful generating methods are not properly linked to the piecemeal process.

42 The urban patterns in the periphery of Ankara are highly determined by the road system making it the first and major binding factor that limits the possibilities of alternative forms in developing sites is the border definition of the road system (Çalışkan, 2012). Vast amount of land in the peripheries is transformed by larger scale plans, however, a significant number of sites have been developed separately leading to no outer morphological reference within the urban context.

Because building blocks are defined by a close loop of roads, the street system which works as a boundary not free from the road surface has a negative effect on possible alternative complexities between buildings (Çalışkan, 2012). This makes the road system the basic element for formation of ensembles.

43 (Figure 2.6. Selected areas of lost spaces.)

The diagram above shows the areas that are the focus of this research. All these areas are part of this developed corridor and are selected as areas with problems, previously named as lost spaces.

Some of the selected areas contain the first type of lost space, while some others the second type as described previously. Even though the selected areas are more structured as urban tissues rather than the surrounding areas, they are not singled out to show categorical changes, rather to make the problematic addressed in this research more visible. In addition, strategic solutions will be introduced to make these urban tissues more valuable in terms of visibility.

44 Çayyolu is a district which has resulted from the city’s urban expansion. Urban expansion is linked with matters that require attention: change of land use, energy demand and fossil fuel consumption, modification of the rural surfaces, vegetation, physical and social changes. Planned extension of the city during the last decades of 20^th century has left large amounts of developable areas without serious construction barriers such as natural or topographical ones. The rapid development process in the area has been associated with a series of fragmentary plans, which has resulted into a nonintegrated whole. Additionally, the limited economic situation of Turkish middle class has affected the suburban extension of the city including this district, creating a hybrid type of urban ensembles. Nevertheless, this condition of hybrid urban ensembles has created urban spaces where some of them can be considered as positive, while a number of them can be seen as negative spaces. For these reasons, Çayyolu has been selected as a case study which is worth analyzing and further development proposal are presented for the future.

When analysing the urban morphology of a site, either uniform or mixed pattern formations can be noticed. While analysing different parts of a plan, the uniform patterns are clearer to grasp because the component plants are essentially of same shape and size. In most cases, uniform patterns derive as a result of single effort of design and production coming from the same agent (Kropf, 2017). Many areas, however, appear as a product of several agents with distinctive characteristics which is also the case in Çayyolu. In many cases, these agents are involved in transformations of already existing urban fabric. The scenario of mixed or heterogenous pattern formations includes areas which are often subject to transformations, mainly the oldest settlements of the city because of their constant need for repair, replacement or extension.

45 The fundamental assumption is that areas composed of different regular patterns suggest different phases of growth and transformation of cities. However, regularity does not necessarily assure planning because planned patterns may also look ‘irregular’ (Kropf, 2017). According to Kropf (2017), an urban area can be rational without the need of being uniform if the area includes a characteristic mixture of elements that are included during a period of piecemeal development. In this mixture, some of the elements may be predominant while others exceptional. A common example of mixed type is a specific route which leads into a town that for a long period of time has undergone several piecemeal developments, making the whole process consistent. Another case where significant variation and heterogeneity exists, tissues may still be distinct but at the same time share common characteristics. The common feature can be a specific house type, however, in some areas semi-detached or terraced houses may also exist.

46 (Figure 2.7. Existing Street Network producing lost spaces.)

47 The diagram above shows a number of selected areas of similar scale that have an already established street network. The street network is different from site to site; some of them being more regular street patterns than others. In any case, isovist analysis will be applied to each site to assess the quality of their open spaces regarding their visibility level. In the less regular or less established street networks we can see a larger number of areas that are left over spaces, while in the more established street patterns there are central spaces that should function as areas of social interaction.

(Figure 2.8. Type I and Type II lost spaces.)

48 The diagram above show both types of lost spaces which are mentioned previously. The first type of lost areas which corresponds to ‘left-over’ can be seen in different parts of Çayyolu district.

There are other pieces of land which are totally undeveloped, however, the ones selected are in close proximity to roads or streets. They can be found in between two developed urban settlements and still have no function. They can be found near road junctions which usually is a valuable piece of land, and still have no contribution to the overall urban environments. The selected areas do not have major topographical obstacles, which means that they could have been developed.

The second type of these lost spaces that one can come across in Çayyolu differs from the first type, but both of their contributuons to the whole urban tissue can be improved. The second type is the central core of the residential develompents there, however, it has few positive contribution to the residents since these spaces are not being utilized at their full potential. They are usually in a smaller scale and are surrounded by residential units, whereas the first type can be found anywhere with varying scales but still in valuable land areas. The identification of both these types of spaces in Çayyolu district is important because it will later be linked with their visibility analysis. The analysis will show that the proposed public spaces are not visually accessible by the majority of surrounding residential units, making these spaces less utilized.

49